fledge-mcp
The Fledge MCP Server connects Fledge functionalities to Cursor AI, enabling natural language interactions with Fledge instances. It offers tools for data management, service control, and advanced AI-assisted features, with options for secure communication using API key authentication.
Fledge MCP Server
This project provides a Model Context Protocol (MCP) server designed to integrate Fledge functionality with Cursor AI. It enables interaction with Fledge instances through natural language commands. Key features include data access, service control, and AI-assisted UI generation. The server can be installed and secured with API key authentication for API calls and is intended for production deployment behind a reverse proxy. For scalability, it can be deployed on platforms like Smithery.ai.
Prerequisites
- Fledge installed locally or accessible via API
- Cursor AI installed
- Python 3.8+
Running the Server
Make sure Fledge is running, then start the MCP server. For secure operations, use API key authentication.
Available Tools
- Data Access: Fetch sensor data, list sensors
- Service Control: Start or stop Fledge services
- AI-Assisted Features: Generate mock sensor data
Production Considerations
- Use HTTPS
- Implement robust authentication like JWT or OAuth
- Rate limiting
- Persistent data storage
JSON-RPC Protocol Support
The server uses JSON-RPC 2.0 over WebSocket, supporting methods like initialize and tools/call.